I am currently working with the 2023 5-Year ACS data and just ran into an issue where I was getting NA values for industry in my tables that used a crosswalk with the 2013-2021 IND codes. On inspection I found that was because the IND variable now uses the 2022-2027 codes.
My question is whether I need to adjust my R code to account for different IND values for the observations in my dataset where MULTYEAR <= 2021, or if all observations in the 5-year data set have the same IND codes. If I do need to create a bifurcated system for IND, is there a crosswalk I can use to compare the old codes to the new ones?
Thanks in advance!
In multi-year samples, IND codes are usually based on the latest year included in the sample. In the 2023 5-year ACS sample, IND codes use the codes from 2023-2027. Since within the multi-year sample, the same industry classification scheme is used for all years, you don’t need to make any adjustments. The PUMS code lists from the Census Bureau for each 5-year ACS sample include more detailed information about how industry was coded and how older industry classification schemes were bridged into the newer ones. Here is the 2019-2023 ACS PUMS code list. You may also be interested in the variable IND1990, which standardizes industry codes from all years from 1950 and onward into the 1990 industry classification scheme.